The present invention relates to a method and a system for (trace) gas analysis.
It is common knowledge that gases may be identified by their absorption spectrum, with the concentration (K) of the gas being derived from the extinction (E) of a certain wavelength which is characteristic of the gas whose presence is to be detected.
In a known system for flue-gas analysis (applicant's own LORA-1 system) a diode laser is used which can be continuously tuned completely and whose output beam, which is set to a known absorption line of a gas to be detected, is passed through an analyzer while the concentration of the measuring gas in the analyzer is attenuated correspondingly. The magnitude of the concentration of the gas under analysis can be derived from such an attenuation of intensity or such an extinction whenever there is no cross-response to the other gases in the analyzer at the measured wavelength and when, moreover, prior to the beginning of a measuring cycle, a calibration value will have been generated (by introduction of a defined measuring gas concentration). As such a calibration with a gas of a known composition is possible only if the probing section is established within an analyzer cell, flue-gas must be passed through this analyzer cell, e.g. when flue gases emitted from a power plant are to be measured, with the maximum rate of measurement being dependent on the rate of purging the analyzer cell so that it is within the range of several seconds per measurement. Apart from the problem of such comparatively low measuring rate, the known system involves the additional problem that prior to the beginning of the measurement sequence (following the shut-down of the system) calibration is required first, which is a very troublesome and complicated, labor-consuming operation, in addition to the necessity of an analyzer cell with its known drawbacks.